<abstract><p>This systematic review aims to investigate recent developments in the area of arc fault detection. The rising demand for electricity and concomitant expansion of energy systems has resulted in a heightened risk of arc faults and the likelihood of related fires, presenting a matter of considerable concern. To address this challenge, this review focuses on the role of artificial intelligence (AI) in arc fault detection, with the objective of illuminating its advantages and identifying current limitations. Through a meticulous literature selection process, a total of 63 articles were included in the final analysis. The findings of this review suggest that AI plays a significant role in enhancing the accuracy and speed of detection and allowing for customization to specific types of faults in arc fault detection. Simultaneously, three major challenges were also identified, including missed and false detections, the restricted application of neural networks and the paucity of relevant data. In conclusion, AI has exhibited tremendous potential for transforming the field of arc fault detection and holds substantial promise for enhancing electrical safety.</p></abstract>

No abstract

No abstract

No abstract

Tax audit is an important part of the tax collection and management system, which directly affects the economic interests of the country and taxpayers. Therefore, reducing the enforcement risk in tax audit is crucial to continuously improve the tax collection and management system. Recently, the research of using deep learning to classify Chinese tax audit data to achieve this goal has attracted much attention. Inspired by BERT, this paper proposes a syntactic enhancement BERT (SE-BERT). It can improve BERT’s text understanding ability by learning input features and grammatical structure of text from text content and location embeddings. In addition, we weight the word importance calculated by TF-IDF with SE-BERT to improve the ability of recognizing local salient features. Through comparative experiments on our Chinese tax audit dataset, our method achieves better performance.